GR

Granite Embedding 97m Multilingual R2

GraniteGGUF by xnohatUpdated 2026-07-13
Download on Hugging Face
Parameters97M
Context32K
Layers12
Architecturemodern-bert
Smallest quant0.1 GB
Quick answer

Granite Embedding 97m Multilingual R2 needs about 0.8 GB of memory at Q4_K_M (a 0.1 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 6 quants, from 0.1 GB (Q3_K_L) to 0.2 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 0.6 GB.

Granite Embedding 97M Multilingual R2 – 384‑dimensional embeddings for 200+ languages

Granite Embedding 97M Multilingual R2 is a 97 billion‑parameter dense embedding model. It produces 384‑dimensional vectors with a context length of 32,768 tokens. The model supports 200+ languages and is trained under permissive licenses.

Good at

  • High‑quality multilingual embeddings
  • Long‑document retrieval
  • Code retrieval

Limits

  • No vision support
  • Requires 8 k context for memory estimates
  • Apache‑2.0 license

Best for

  • Embedding generation for search engines
  • Multilingual information retrieval
  • Code search

Who it’s for. Developers building multilingual search or retrieval systems that need compact embeddings. Researchers needing a lightweight model for long‑context embedding tasks.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_L0.1 GB 0.8 GB
Fits Get file →
Q4_K_M0.1 GB 0.8 GB
Fits Get file →
Q5_K_M0.1 GB 0.8 GB
Fits Get file →
Q6_K0.1 GB 0.8 GB
Fits Get file →
Q8_00.1 GB 0.8 GB
Fits Get file →
BF160.2 GB 0.9 GB
Fits Get file →

For 8 GB VRAM, Q3_K_L, Q4_K_M, Q5_K_M, or Q6_K are suitable, each about 0.8 GB memory at 8 k context. Q8_0 is recommended for best quality with a 0.12 GB file. BF16 uses 0.9 GB memory and 0.21 GB file size.

What GPU do you need for Granite Embedding 97m Multilingual R2?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: the best quant that fits is BF16 (0.2 GB file, about 0.9 GB in use).

Which Granite Embedding 97m Multilingual R2 quant should you choose?

Q4_K_M is the usual starting point for Granite Embedding 97m Multilingual R2: it balances file size and quality well. Here it is a 0.1 GB file.

The Q8_0 file is 0.1 GB, about 1.1 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is Q3_K_L at 0.1 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturemodern-bert
Parameters97M
Layers12
Trained context32,768 tokens
KV cache (fp16)18 KB per token

Source

Quantized byxnohat
FamilyGranite
Licenseapache-2.0
Downloads (HF)608
Last updated2026-07-13

Frequently asked questions about Granite Embedding 97m Multilingual R2

How much VRAM does Granite Embedding 97m Multilingual R2 need?

About 0.8 GB at Q4_K_M with an 8K context window. That is the 0.1 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Granite Embedding 97m Multilingual R2 on an 8 GB GPU?

Yes, with BF16: the 0.2 GB file needs about 0.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Granite Embedding 97m Multilingual R2 GGUF?

It depends on the quant: Q3_K_L is 0.1 GB, Q4_K_M is 0.1 GB, Q5_K_M is 0.1 GB, Q6_K is 0.1 GB, Q8_0 is 0.1 GB, BF16 is 0.2 GB.

What context length does Granite Embedding 97m Multilingual R2 support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 0.2 GB at 8K and 0.6 GB at 32K.

Where can I download Granite Embedding 97m Multilingual R2 GGUF?

The files are on Hugging Face in the xnohat/granite-embedding-97m-multilingual-r2-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Granite Embedding 97m Multilingual R2 use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run Granite Embedding 97M on 8 GB VRAM?

Yes, the Q3_K_L to Q6_K quantizations use about 0.8 GB memory at 8 k context, fitting within 8 GB VRAM.

Does the model support image inputs?

No, the model only processes raw text and does not support vision.

What pooling type does the GGUF use?

The GGUF files use CLS pooling, which is baked into the model and should not be changed.

Other GGUFs of granite-embedding-97m-multilingual-r2

Similar models

Browse all →

More Granite models

All Granite →

Other tiny models

All tiny models →